> On 18 Mar 2026, at 19:39, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Daniel Gustafsson <daniel@yesql.se> writes:
>> I've done more testing now and I can only reproduce this with downgraded
>> versions of OpenSSL 1.1.1, when running the latest 1.1.1x the error goes away
>> and the error is reported as expected. Can you try to upgrade your machine,
>> which I assume isn't running bleeding edge 1.1.1 by the sounds of it.
>
> Nope, it was still on 1.1.1a.
Wow, that was almost an Antiques Roadshow material candidate =)
> I've now updated it to 3.0.19,
> which appears to be the oldest available-to-the-public supported
> version. I doubt that continuing to test 1.1.1-anything is really
> useful.
I don't disagree, 1.1.1 went EOL a long time ago, but since we still support
1.1.1 we need to test it somewhere. I believe we have sufficient coverage
though, and I personally test on 1.1.1 through 3.6 (and will start testing 4
soon).
> ... and longfin is now green.
Great, thanks!
--
Daniel Gustafsson
